Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact

Transcription

1 Antioch University AURA - Antioch University Repository and Archive Dissertations & Theses Student & Alumni Scholarship, including Dissertations & Theses 2014 Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact Matthew Dean Stevens Antioch University - New England Follow this and additional works at: Recommended Citation Stevens, Matthew Dean, "Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact" (2014). Dissertations & Theses This Dissertation is brought to you for free and open access by the Student & Alumni Scholarship, including Dissertations & Theses at AURA - Antioch University Repository and Archive. It has been accepted for inclusion in Dissertations & Theses by an authorized administrator of AURA - Antioch University Repository and Archive. For more information, please contact dpenrose@antioch.edu, wmcgrath@antioch.edu.

2 Running head: PREDICTING INTENSITY OF CONTACT Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact by Matthew Dean Stevens B.A., University of Massachusetts-Amherst, 2001 M.S., Antioch University-New England, 2007 DISSERTATION Submitted in partial fulfillment for the degree of Doctor of Psychology in the Department of Clinical Psychology at Antioch University New England, 2014 Keene, New Hampshire

3 PREDICTING INTENSITY OF CONTACT ii Department of Clinical Psychology DISSERTATION COMMITTEE PAGE The undersigned have examined the dissertation entitled: MAINTAINING BBBS MENTORING RELATIONSHIPS: EXPLORING PREDICTORS OF INTENSITY OF CONTACT presented on July 24, 2014 by Matthew Dean Stevens Candidate for the degree of Doctor of Psychology and hereby certify that it is accepted*. Dissertation Committee Chairperson: George Tremblay, PhD Dissertation Committee members: Martha B. Straus, PhD William Slammon, PhD Accepted by the Department of Clinical Psychology Chairperson Kathi A. Borden, PhD on 7/24/14 * Signatures are on file with the Registrar s Office at Antioch University New England.

4 PREDICTING INTENSITY OF CONTACT iii Acknowledgements First, I would like to express sincere gratitude to my dissertation chair, George Tremblay, Ph.D. for his continuous support, patience, guidance, and knowledge throughout this process. I am grateful for his persistence in working with me to get through many challenging obstacles. I would also like to thank Dr. Tremblay for all his help and support during both qualifying exams. In addition to my advisor, I would like to thank the rest of my dissertation committee: Martha Straus, Ph.D. and William Slammon, Ph.D. for their encouragement, insightful comments, and hard questions. I also want to thank all of the Big Brothers and Big Sisters agencies that supported this research and distributed the survey to eligible volunteers. I also want to thank my parents and sister whose love and support growing up gave me the courage to set high goals and the confidence to believe I could achieve them. Last but not least, I would like express sincere thanks and gratitude to my wife who remained extremely supportive and patient throughout this long process. She endured the highs and lows that occurred and did whatever she could to allow me to complete this project.

5 PREDICTING INTENSITY OF CONTACT iii Table of Contents List of Tables... vi List of Figures... vii Abstract...1 Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact...3 Goals of Research...5 Context of Research...7 An Overview of Mentoring Research...8 The Conception, History, and Evolution of Mentoring...8 Theories of Volunteerism and Mentoring...10 Psychosocial theories of mentoring...11 Sociological theories of mentoring...13 Functional theories of mentoring...14 The Role of a Youth Mentor...15 Sustaining Youth Mentoring Relationships and Effects of Adult and Youth...16 Importance of retaining mentors...16 Predictors of persistence in mentoring...17 The Challenges Volunteer Organizations are Facing...19 Summary of the Research on Mentoring...22 Overview of the Study...23 Research Questions and Hypotheses...24 Methods...25 Participants...25 Procedure...26 Recruitment...26 Measure...27 Description and Measurement of Variables...29 Results...31 Data Preparation...31 Predicting Intensity of Contact...33

6 PREDICTING INTENSITY OF CONTACT iv Using a Discrepancy between Initial Expectations and Experience to predict Intensity of Contact...33 Examining the relationship between Frequency of Match Support, Expectation-Experience Discrepancy, and Intensity of Contact...36 Analyzing the effect of Initial Expectations on Intensity of Contact...38 Discussion...39 Summary of Results...39 Explanation of Findings...40 Using a Discrepancy between Initial Expectations and Experience to predict Intensity of Contact...40 Examining the relationship between Frequency of Match Support, Expectation-Experience Discrepancy, and Intensity of Contact...42 Analyzing the effect of Initial Expectations on Intensity of Contact...42 Limitations of the Study...42 Comparisons of Results with Past Research...43 Areas for Future Research...43 Conclusions...45 References...47 Appendices Appendix A: Permissions...53 Appendix B: Recruitment Appendix C: Participant Informed Consent...56 Appendix D: Sweepstakes Drawing...58 Appendix E: Distributions for Primary Research Variables...59 Appendix F: Frequency Data for Initial Expectations for the Six VFI Subscales...61 Appendix G: Frequency Data for Experience Rating Score for the Six VFI Subscales...64 Appendix H: Frequency Data of Discrepancy Scores for the Six VFI Subscales...67

7 PREDICTING INTENSITY OF CONTACT v List of Tables Table 1. Description of Variables, Hypotheses, and Analyses...31 Table 2. Descriptive Statistics for Focal Variables...33 Table 3. Using a Discrepancy Score to Predict Intensity of Contact...36 Table 4. Two Level Match Support and Associated Discrepancy Scores and Hours of Face-to-Face Contact...37 Table 5. Correlation Matrix for Frequency of Match Support, Discrepancy Score, and Intensity of Contact...38 Table 6. Correlations between Initial Expectations and Intensity of Contact...39

8 PREDICTING INTENSITY OF CONTACT vi List of Figures Figure 1. Scatter plot of Discrepancy Score and Intensity of Contact. 35

9 PREDICTING INTENSITY OF CONTACT 1 Abstract This study surveyed a group of adult mentors at Big Brothers and Big Sisters program (BBBS) sites to examine variables that affect an adult volunteer s intensity of contact in youth mentoring. This study attempted to expand on research conducted by Clary et al. (1998) and Madia and Lutz (2004). In this cross-sectional design, participants were administered the Volunteer Functions Inventory to learn about their expectations and experiences as mentors. Adult mentors reported on the number of hours of face-to-face contact with their mentee during the previous month. The primary hypothesis is that congruence between initial expectations and actual experience of adult mentors will predict the intensity of contact with a child or adolescent. A regression analysis was used to assess the relationship between volunteer engagement and the discrepancy between initial expectations and actual experience. Secondary objectives involved examining how the relationship between the expectations-experience discrepancy score and intensity of contact are affected by the frequency of match support provided by BBBS case managers. Additional analyses included assessing the relationship between intensity of contact and specific motivational factors, as measured by Clary et al. s Volunteer Functions Inventory. Results of all analyses revealed that these variables are not reliable predictors of current volunteer engagement; however, the data did provide a few helpful conclusions. The results suggest that the effects of a discrepancy between a volunteer s expectations and experience on intensity of contact are minimal; however, the wide range in hours of face-to-face contact suggests that some other variables are affecting intensity of contact. Additionally, the frequency of match support, a service provided to all BBBS volunteers, showed no association with the discrepancy score or intensity of contact. It is possible that that restricted range in discrepancy scores masked the effects provided by match support. Lastly, areas for future research include exploring other

10 PREDICTING INTENSITY OF CONTACT 2 possible determinants of intensity of contact, using qualitative methods for identifying the effective aspects of match support, and examining if there is a minimal frequency threshold for youths to experience benefits. Keywords: Intensity of Contact, Mentoring, Big Brothers Big Sisters

11 PREDICTING INTENSITY OF CONTACT 3 Maintaining BBBS Mentoring Relationships: Exploring Predictors of Intensity of Contact Mentoring is a popular and widely used intervention to improve the lives of young people. Research indicates that over 4,500 agencies across the United States provide mentoring services to youth (Rhodes, 2002). Big Brothers and Big Sisters of America (BBBS) is an organization with sites nationwide. The agency matches children and adolescents with adult volunteers who provide support, nurturance, friendship, and guidance to help children and adolescents overcome challenges, and achieve their highest potential. A large body of research has demonstrated that BBBS mentoring programs help young people overcome personal and social obstacles through the development of new competencies such as leadership, problem solving, and other life skills (McDonald-Hart, 2002; Westhues, Clarke, Watton, & St. Claire-Smith, 2001). Other benefits youth receive from mentoring include improvements in self-esteem and academic achievement, lower dropout rates, and decreases in truant and delinquent behaviors (Tierney, Grossman, & Resch, 1995). The popularity of youth mentoring is fueled by research on resilience, which validates the public s perceptions that at-risk youth can experience positive developmental outcomes if they are connected to just one supportive adult (Dubois & Karcher, 2005; Rhodes, 2002). Many youth do not readily find older, supportive adults within or outside of their homes. Shifting marital patterns, overcrowded schools, and loss of community cohesiveness have significantly reduced the availability of caring adults and restricted their opportunities for informal contact with youth (Rhodes, 2002). The proportion of single-parent families more than doubled from 1980 (18% of all families) to 2008 (40%) approximately 1 in 3 children in the US reside in a single-parent family (U.S. Census, 2012). Manufacturing jobs that provided families economic

12 PREDICTING INTENSITY OF CONTACT 4 stability and supported two -parent families have given way to low-paying service jobs and unemployment (Rhodes, 2002). These changes within families and communities coincided with rapidly escalating rates of juvenile arrest and violent crime during the 1980s and 1990s (Rhodes, 2002). Although juvenile violent crime has trended downward during the last decade (Puzzanchera & Adams, 2011), it still appears that many youths do not have frequent contact with positive role models. Youth mentoring agencies try to provide youth with positive role models; however, they have trouble recruiting sufficient numbers of mentors to meet the demand, and sustaining relationships over time (National Partnership for Mentoring, 2005). With recent decreases in government and other sources of public funding, human service agencies and community development organizations are increasingly dependent on volunteer involvement to carry out their missions (Bussell & Forbes, 2002). At the same time, volunteer organizations have struggled to meet the increased demands for volunteer work because adult mentors have less time to commit to volunteering due to changes in families, an increase in individuals pursuing graduate degrees, and lack of stable communities (Rhodes, Bogat, Roffman, Edelman, & Galasso, 2002). Also, many more young adults who are in a fertile developmental period for mentoring attend graduate school to meet the demands of an increasingly competitive job market, which results in less free time. Last, the shrinking pool of potential volunteers has caused many children and adolescents who are referred to mentoring programs to either end up on waiting lists for up to a year before being matched to an adult volunteer or are denied services because over-subscribed mentoring programs have created restrictive eligibility criteria to shorten wait lists (Rhodes, 2002). As if the mismatch of supply and demand for mentors weren t sufficiently challenging, almost half of newly formed mentoring relationships terminate within the first few months. This

13 PREDICTING INTENSITY OF CONTACT 5 is an alarming statistic because the duration of the mentoring relationship consistently emerges as an important moderator of outcomes. Generally, youth involved in longer-lasting relationships tend to experience better outcomes. They experience gains in self-worth, do better in school, get along better with their parent(s), and exhibit lower rates of substance abuse when compared to youth in relationships that terminate after a few months (Rhodes & Grossman, 2002). Consequently, researchers have devoted considerable time to exploring how a youth and an adult mentor develop a meaningful relationship. Common themes across all successful types of relationships include a mutual sense of connection, trust, empathy, and respect (Rhodes, 2002). In addition, the bond that is formed tends to become stronger as the duration of the relationship increases (Rhodes, 2002). Not surprisingly, developing an effective relationship with a mentee requires the mentor to invest considerable time, effort, and patience. Furthermore, social exchange theorists posit that the survival or demise of all relationships depends largely on the rewards of the relationship outweighing its costs. Volunteers and youths can be drawn to the relationship in anticipation of imagined rewards, but real rewards are needed to develop and sustain it. In addition, the rewards each member in the relationship looks for can change over time. These findings require that mentoring agencies develop strategies to both identify adults more likely to persist and to monitor satisfaction. Therefore, this study examined variables that affect volunteer recruitment and retention to aid organizations like BBBS so that they might improve practices and procedures that support both recruitment and retention of motivated mentors. Goals of Research My interest in studying the intensity of contact of adult volunteers is generated by my own involvement as a Big Brother in the Hampden County Office of Big Brothers and Big

14 PREDICTING INTENSITY OF CONTACT 6 Sisters community-based program, located in Springfield, MA. This study was conducted to help agencies improve recruitment and retention strategies by exploring predictors of the intensity of the volunteer s contact with his or her mentee. Among the many factors that might contribute to a mentor s contact, and one over which mentoring agencies potentially have some influence, is the fit between a mentor s expectations and his or her actual experience with the agency. There is reason to suspect that experience that violates a mentor s expectations is more likely to lead to early attrition of mentors. Minimizing volunteer attrition is important because mentoring relationships that terminate within the first six months are less beneficial and can even have negative effects on the youth. Furthermore, research suggests that most youth are matched for at least six months before meaningful changes are observed. Consequently, it would seem that the discrepancy between initial expectations and actual experience would begin to get smaller for adult volunteers once they can observe or notice positive changes in the youth. Therefore, it is believed that congruence between initial expectations and actual experience of adult mentors will predict their intensity of contact as volunteers with BBBS community-based mentoring. A second goal of this research includes examining how the frequency of match support provided by BBBS case managers affects the discrepancy between expectations and experience and intensity of contact. Currently, BBBS agencies decrease the intensity of match support at six months post match, despite the significant attrition that occurs. Data from this analysis could identify volunteers still vulnerable to dropping out and help case managers to focus their support. The last goal involves assessing a relationship between the intensity of contact and the volunteer s specific motivational factors, as measured by Clary et al. s (1998) Volunteer

15 PREDICTING INTENSITY OF CONTACT 7 Functions Inventory. The results of this analysis could help BBBS agencies identify expectations that increase the risk of premature termination. Context of Research BBBS is a mentoring program that was started in 1904 by Ernest Coulter, a New York City court clerk. Coulter started the program in response to an influx in the number of male youth becoming involved with the courts. He believed that caring adults could help many of these kids stay out of trouble, and he set out to find volunteers. That marked the beginning of the Big Brothers movement. At around the same time, the members of a group called Ladies of Charity started their own mentoring program for girls who had come through the New York Children s Court. That group would later become Catholic Big Sisters. Both groups functioned independently until 1977, when Big Brothers Association and Big Sisters International joined forces to become Big Brothers Big Sisters of America (Big Brothers Big Sisters, 2014a). Big Brothers and Big Sisters is the oldest and largest youth mentoring organization in the United States, currently serving children ages 6 to 18, and operating in all 50 states and 12 countries around the world. In the past few years, BBBS has expanded its programs to serve more youth by offering school-based mentoring programs such as Lunch Buddies and after school programming. In addition, the organization recognizes the diversity of our youth and is working to create specialized mentoring services. BBBS has created specialized mentoring services for African American, Hispanic, and Native American youths as well as for children with an incarcerated parent or a parent in the military (Big Brothers Big Sisters, 2014b). More than 100 years after its inception, Big Brothers and Big Sisters remains true to its mission of bringing caring role models into the lives of all children.

16 PREDICTING INTENSITY OF CONTACT 8 An Overview of Mentoring Research Mentoring is a long-standing intervention that has been used to improve the lives of young people through guidance and support. Researchers have studied many concepts to improve its use and effectiveness for those who need it. The literature on mentoring provides information on factors that affect recruitment and retention of mentors, theories that explain engagement and commitment to volunteerism and mentoring, the role of a mentor, what motivates people to become mentors, ways to assess the motives of mentors, and the benefits adult volunteers receive from mentoring. All of these topics will be reviewed in the sections that follow. The Conception, History and Evolution of Mentoring The term mentor owes its origin to the epic Greek poem, The Odyssey, which was attributed to Homer in approximately the 8 th century BCE (McDonald-Hart, 2002). According to this legend, Athena, the goddess of wisdom, took the form of Mentor and provided support and guidance to Odysseus s son, Telemachus, while Odysseus was away fighting the Trojan War. Athena connected Telemachus to people and resources that could guide his development. Since its initial reference in this Greek legend, the concept of mentoring has continued to grow and evolve, with researchers offering a variety of conceptualizations (Dubois & Karcher, 2005). Current research indicates the concept of mentoring consists of three core conceptual elements. First, a mentor is a person with greater experience or wisdom than the mentee. Second, the mentor provides support, guidance, and instruction to help promote the development of the mentee. Last, the mentor and mentee form an emotional bond characterized by a sense of trust, which creates safety and enables the mentee to take risks and grow (Dubois & Karcher, 2005). These three core concepts provide the foundation of youth mentoring for agencies like BBBS.

17 PREDICTING INTENSITY OF CONTACT 9 Initially, mentoring was perceived as a strategy through which those in powerful positions groomed a mentee. In the working world, mentoring is defined as an intense interpersonal exchange between an experienced senior employee mentor and a less experienced junior employee mentee in which the mentor provides support, direction, and feedback regarding the mentee s career plans and personal development (McDonald-Hart, 2002, p. 14). Indeed, much of the early research on mentoring focused primarily on these career functions. More recently, the concept of mentoring has evolved to include psychosocial and role modeling functions (Brown-Wright, Dubick, & Newman, 1997). In role model mentoring, the mentor models positive qualities and behaviors that the mentee can learn and emulate (McDonald-Hart, 2000). Psychosocial mentoring is conceptualized as a relationship focused on improving a mentee s self-worth, competence, identity, and effectiveness through friendship, acceptance, and counseling (McDonald-Hart, 2002, p. 14). Philip and Hendry (2000) classified the mentoring provided to children and adolescents into five types: (a) classic mentoring, (b) individual-team mentoring, (c) friend-to-friend mentoring, (d) peer group mentoring, and (e) long-term relationship mentoring with risk taking adults. Classic mentoring occurs when a youth is paired with an older individual who offers the youth connection, guidance, support, and access to resources. Individual-team mentoring occurs when a group of people searches for a person or persons to provide them with support, advice, and knowledge to overcome challenges. An example of individual-team mentoring would be the formation of a church youth group or some other specialized youth group. The difference between classic mentoring and individual-team mentoring is that in individual-team mentoring, the mentor is not necessarily older than the group members but is deemed to have some special knowledge, wisdom, or insight. A friend-to-friend mentoring relationship routinely happens

18 PREDICTING INTENSITY OF CONTACT 10 among youth during adolescence, a period of development during which friends become more influential and are sought for advice and support. Peer group mentoring is described as a group that provides support to help an individual overcome challenging social situations. Alcoholics Anonymous (AA) is an example of peer group mentoring. Last, long-term relationship mentoring with risk taking adults is comparable to classic mentoring, but differs in that the mentor and mentee are paired because they share a similar background such as rebellion or challenging of authority. It seems this type of mentoring is less prevalent because many people fear that pairing an at-risk youth with a previously troubled adult could reinforce the youth s problematic behavior. Of these five types of mentoring, classic mentoring has been the most heavily researched type, and is the focus of this study. Theories of Volunteerism and Mentoring Researchers have utilized both psychological and sociological theories to explain why people volunteer. The initial focus of these theories was on individual characteristics and traits associated with volunteering. The focus then shifted to developing stages or processes that explain engagement and persistence in volunteer work. Now, theorists are focused on understanding a volunteer s motives. Psychological theories of mentoring. Some of the psychologically based research on volunteerism identified several personality traits that influence involvement and commitment to mentoring such as empathy, concern for the rights of others, a need to help others, and a desire to be useful and feel needed (Anderson & Moore, 1978; Penner, 2002; Penner & Finkelstein, 1998; Penner et al., 1995). Penner et al. (1995) developed the Prosocial Personality Battery to assess these traits, which have been shown to predict volunteering for multiple organizations, longer periods of service, and greater overall time commitment. Allen (2003) administered the

19 PREDICTING INTENSITY OF CONTACT 11 Prosocial Personality Battery (PSB) to a group of 391 mentors and non-mentors in a variety of settings and reported that prosocial dispositions are also associated with the propensity to mentor others. For example, research indicates people who score high on helpfulness often engage in volunteer work (Allen, 2003). Furthermore, Allen explained that mentors who reported being motivated by intrinsic satisfaction and had high scores on the Other-oriented Empathy scale were more likely to engage in psychosocial mentoring. Last, Rhodes and Grossman (2002) concluded that agencies must be aware of these factors so staff selects only those individuals who are most likely to make the commitment to mentoring. Research shows that both participation and motivation for an individual s engagement in a volunteer activity is influenced by demographic variables such as employment status, educational level, socioeconomic status, racial or ethnic identity, and developmental stage. Employed, highly educated individuals tend to volunteer more often and tend to be more motivated by opportunities for self-fulfillment, while unemployed, less educated individuals tend to volunteer less often and tend to engage in volunteer work to feel useful or occupy spare time (Anderson & Moore, 1978; Wilson, 2000). Individuals from racial and ethnic minority groups volunteer less, although Wilson has shown that when income and education are controlled for, racial and ethnic identity no longer predict rates of volunteerism. Larkin, Sadler, and Mahler (2005) illustrated how an individual s developmental life stage influences motives. These researchers studied a group of older adult volunteers who mentor at-risk youth and found that these volunteers were motivated by the developmental concept of generativity sharing one s accumulated good fortune or wisdom with others (Larkin et al., 2005). Some other research has also found sex differences affect an adult s expectations of youth mentoring. Women volunteer to feel useful and occupy spare time and are encouraged to engage in volunteer work that shows

20 PREDICTING INTENSITY OF CONTACT 12 sympathy (Anderson & Moore, 1978; Eagly & Crowley, 1986; Stukas & Tanti, 2005; Wilson, 2000). Conversely, men expect volunteering to provide self-fulfillment and enhance personal development and are encouraged to volunteer for activities that require strength and heroism (Anderson & Moore, 1978; Eagly & Crowley, 1986; Stukas & Tanti, 2005; Wilson, 2000). For example, more men than women usually volunteer to coach youth sports or referee games. Therefore, to improve recruitment and retention of male mentors, mentoring agencies have created sports buddy programs designed to have a male mentor play sports with their mentee (Bogat & Liang, 2005). This research is helpful because it describes characteristics associated with volunteerism, but it doesn t tell us much about a volunteer s expectations and the underlying processes that account for whether volunteers will persist. Psychologically based research has evolved toward theoretical frameworks designed to identify the developmental stages through which a volunteer progresses during their service. Omoto and Snyder (1995) described a conceptual framework that focused on the antecedents that trigger engagement, the experiences that are encountered during the volunteer role, and the consequences that influence the duration of a volunteer s service. The antecedent stage includes variables that trigger engagement in volunteer work, including personality attributes that predispose people to become involved in helping relationships, the individual s personal, social, and motivational needs, as well as aspects of the person s life and lifestyle that promote engagement in volunteer work. The experience stage describes volunteers who are involved in activities that satisfy their motives, needs, and goals, and volunteers who are more likely to persist as volunteers. The consequences stage includes both the duration of the volunteer s service and the benefits from volunteering (Omoto & Synder, 1995). This model is helpful because it explains the effects of antecedent and experience factors on the intensity of the

21 PREDICTING INTENSITY OF CONTACT 13 volunteer s contact with a youth. Moreover, it represents an improvement from earlier theories because it integrates multiple variables to explain engagement and intensity of contact in volunteer work. Another benefit of the stage model is that it provides an outline of a volunteer s progression from initial engagement to prolonged involvement and can help case managers both anticipate and resolve potential problems that could affect volunteer contact. However, while Omoto and Snyder offer conclusions about the volunteer s expressed intentions to continue their service, they do not provide any information about how current engagement is affected by the degree of congruence between expectations and experience. Sociological theories of mentoring. Social learning theory provides a framework for understanding an individual s expectations about volunteering through various psychosocial influences on behavior, including the volunteer s history, environment, expectations, observational learning, reinforcement, and self-efficacy. For example, in one study of volunteer peer educators on a college campus, Klein and Sondag (1994) found that life experiences, efficacy beliefs, and social reinforcement were important determinants of volunteering. In addition, Klein and Sondag also reported the role of expectations was less clear because many participants reported not having any clear expectations in advance of volunteering. Other sociologically based research has revealed that people who incorporate being a volunteer into their overall identity are more likely to engage and persist as volunteers. Grube and Piliavin (2000) created the concept of role identities and defined them as components of the self that correspond to the social roles we play (p. 1108). Among a sample of volunteers involved with the American Cancer Society, Grube and Piliavin found that perceived expectations of significant others had the strongest influence on the development of a volunteer role identity, followed by organizational variables such as feeling valued by the organization,

22 PREDICTING INTENSITY OF CONTACT 14 prestige, and being connected to volunteers that they liked. Grube and Piliavin argued that organizations could improve volunteer retention by exerting social pressure on new volunteers, quickly engaging them in desired behaviors, and thus promoting role identity (p. 1117). These results provided clear evidence that an organization s policies and practices can significantly affect an individual s persistence as a volunteer (Dubois & Karcher, 2005; Eisner et al., 2009). Functional theories of mentoring. If social learning theory and role identity theory help us identify common factors associated with volunteering behavior, then functional theories help identify the psychological consequences generated by volunteerism. Functional analysis, in this context, examines the motives that are satisfied, the needs that are met, and the goals that are reached when a person volunteers. A critical aspect of functional analysis is that it recognizes people can possess similar beliefs but engage in the same activity for different reasons (Clary, Snyder, & Ridge, 1992); thus, functional theories recognize the diversity of motives that might guide people toward volunteering. One implication of functional theories is that agencies seeking to recruit volunteers should be concerned with identifying the motives, needs, and goals of their volunteers, and should design and market volunteer roles to satisfy an array of such functions (Clary et al., 1992). For these reasons, this study applied Clary et al. s functional theory in examining factors that predict the intensity of contact in a group of adults who volunteer for Big Brothers and Big Sisters community-based mentoring programs. The Role of a Youth Mentor Mentors are professional helpers who assume a variety of roles such as tutor, sponsor, and confidante, and are expected to model admirable personal traits and professional skills (Wilson & Johnson, 2001). In addition, research has identified some basic expectations applicable to all types of mentors:

23 PREDICTING INTENSITY OF CONTACT 15 Mentors are expected to (a) make a commitment to being consistent and dependable, maintaining a steady presence in the youth s life; (b) recognize that relationships may be fairly one-sided and take responsibility for keeping the relationships alive; (c) respect the youth s viewpoint; (d) involve the youth in deciding how the pair will spend their time together; (e) pay attention to the youth s need for fun; (f) develop and maintain a good relationship with the mentee s families; and (g) seek and utilize the help and advice of program staff. Conversely, mentors are advised (a) not to expect the youth to take on equal responsibility for initiating contact; (b) not to try to teach the mentee a set of values and beliefs that are unlike those the youth is exposed to in the home, (c) not to take on a parental or authoritative role, and (d) not to be solely focused on behavior change. (Sipe, 2002, p. 254) While these roles and responsibilities are all important, the mentor s primary task is to develop a relationship characterized by mutuality, trust, and empathy because it increases the mentee s receptiveness to guidance and advice, which leads to growth in social skills, thinking skills, and emotional functioning (Rhodes, 2005). However, Rhodes (2002) cautions, if a bond does not form, the youth and mentor may disengage from the match before the relationship lasts long enough to have a positive impact on the youth (pp ). These comments highlight the importance of identifying volunteers most likely to persist because developing a relationship characterized by trust, mutuality, and empathy will require considerable time, effort, and patience. In general, researchers have done a good job identifying and describing the role of a youth mentor and the ways they can facilitate positive outcomes in their mentees. Agencies like BBBS have utilized this research to develop training materials for their mentors to help them

24 PREDICTING INTENSITY OF CONTACT 16 provide positive experiences for youth. Unfortunately, staff at many mentoring agencies have high caseloads, and some agencies lack the necessary resources to develop adequate training materials, provide ongoing training and support to their mentors, or both (Rhodes, 2002). An agency s inability to provide training and ongoing support is a primary problem interfering with volunteer retention. The effect of these organizational practices, as well as the obstacles that interfere with volunteer recruitment and retention, are discussed later on in the review. Sustaining Youth Mentoring Relationships and the Effects on Adult and Youth Importance of retaining mentors. Researchers have extensively studied the outcomes that both mentors and mentees experience from their involvement in mentoring programs. Retaining mentors is vital because studies have shown that the adult volunteer and youth need to maintain a relationship for approximately 6 12 months for meaningful changes to occur (Rhodes, 2002); unfortunately, only 38% of mentoring relationships last one year (MENTOR, 2005). Conversely, youth involved in matches lasting fewer than six months usually experience negative outcomes. Studies indicate that youth involved in matches lasting fewer than six months experienced an increase in alcohol use, while youth involved in relationships that terminated within the first three months experienced significant drops in feelings of self-worth and perceived academic competence (Rhodes & Grossman, 2002). Therefore, agencies need to improve their ability to decrease premature terminations and enable youth to experience the benefits provided by prolonged relationships. Predictors of persistence in mentoring. A common finding from research on longevity of mentoring relationships is that the degree of perceived similarity between the individuals is a strong predictor of the degree of contact between mentor and mentee (Madia & Lutz, 2004). When people perceive high levels of similarity, for example, they have fewer differences to

25 PREDICTING INTENSITY OF CONTACT 17 resolve and can focus on developing a meaningful, supportive, and productive relationship (Hays & Oxley, 1986). Therefore, BBBS agencies prioritize matching youths with adults who possess similar interests, hobbies, and other characteristics to increase the likelihood of a successful relationship. This degree of similarity can change over time because people adopt new values, interests, or beliefs. Fortunately, these differences can be resolved and the relationship maintained if people are provided with appropriate supervision and support (Hays & Oxley, 1986). Therefore, agencies must place equal emphasis on both their matching process and their support of ongoing mentoring relationships so that youth experience the benefits of sustained relationships. More recent research has examined how the intensity of contact is affected by the similarity between a mentor s initial expectations of his or her role and his or her actual experience. For example, Madia and Lutz (2004) surveyed a group of mentors from a BBBS program to determine to what degree similarity and expectation reality discrepancies predict positive relationship characteristics and one s intention to remain in the relationship. Participants were instructed to complete various personality and interest inventories for both themselves and their mentees. Then they were asked to recall their initial expectations of being a mentor prior to volunteering and then to rate how well their experience fulfilled those expectations. In this study, perceived similarity on personality attributes and interests, with the exception of extroversion, did not provide a strong predictor for the intensity of contact in mentors (Madia & Lutz, 2004). However, and most relevant to this dissertation project, a volunteer s expressed intention to continue to volunteer was stronger when paired with a low

26 PREDICTING INTENSITY OF CONTACT 18 discrepancy score, but became weaker when combined with a high discrepancy score (Madia & Lutz, 2004). Research has also identified predictors of mentor attrition which include fearing that one will be ineffective and a failure, feeling that the youth is not trying to change, and not feeling appreciated by their mentee (Rhodes, 2006). Parental status and lower income are also associated with an elevated risk for premature termination, most likely because decreased financial resources and the demands of raising children significantly hinders a volunteer s ability to maintain contact with the youth (Rhodes & Grossman, 2002). Shifting from characteristics of mentors to characteristics of mentees, adolescents appear harder to mentor than latency age children. In one study, matches involving 13 to 16 year olds were 65% more likely to terminate than matches with 10 to 12 year olds (Rhodes & Grossman, 2002). In addition, the risk for premature termination increases significantly when adult volunteers are matched with youth who were referred for psychological or educational problems, or had sustained emotional, sexual, or physical abuse. Researchers report that premature termination of mentoring relationships involving emotionally or behaviorally troubled youth appears to occur more often because mentoring agencies do not inform volunteers of the potential vulnerabilities and challenges of the youth, resulting in the volunteer feeling burdened by their youth s issues, feeling powerless to help, and feeling overwhelmed by the difficult circumstances or neediness of their mentees (Rhodes, 2002; 2006). Due to these findings, BBBS agencies typically refer youth with severe emotional and behavioral problems to more intensive treatment interventions (Rhodes, 2002). BBBS agencies justify this practice by explaining that their volunteers are not experts in child psychology or crisis management (Rhodes, 2002). Unfortunately, this position is a difficult problem for agencies to deal with because many of the

27 PREDICTING INTENSITY OF CONTACT 19 children referred to their programs have, by definition, inadequate adult presence in their lives, typically resulting in some emotional and behavioral problems. The Challenges Volunteer Organizations are Facing In 2005, the National Mentoring Partnership administered a survey to a nationally representative sample of 1,000 adults, asking whether they were in a formal or informal mentoring relationship. Researchers extrapolated from the sample to the population as a whole and estimated that about three million adults nationwide were involved in formal mentoring relationships with youth, and that 44 million adults were interested in youth mentoring but were not volunteering at the time (National Mentoring Partnership, 2005). The phenomenon of intending but not doing has been referred to as the problem of inaction (Stukas, et al., 2006 Stukas & Tanti, 2005). Agencies try to combat the problem of inaction by attempting to persuade people to volunteer through word of mouth, the use of public service announcements, commercials, print advertisements, and organized presentations by program staff and current adult mentors (Stukas & Tanti, 2005). Unfortunately, these recruitment strategies seem to contribute to the problem of inaction because programs do not publicize the potential vulnerabilities of referred youth and they tend to mislead volunteers about the potential benefits and degree of commitment required (Rhodes, 2002). Given that meaningful change takes time to occur, a mentor may feel that he or she is not making a difference and, thus, experience frustration, disappointment, or both. Therefore, programs tend to attract volunteers who turn out to be unwilling or unable to make the necessary long-term commitment (Rhodes, 2002). Many youth mentoring agencies lack the resources to effectively recruit and retain mentors. Some agencies lack the necessary resources to develop adequate training materials, provide ongoing training and support to their mentors, or both (Rhodes, 2002). The limited

28 PREDICTING INTENSITY OF CONTACT 20 financial resources possessed by many mentoring agencies require that case managers perform many different tasks. In some agencies case managers are responsible for interviewing volunteers and youth, conducting background checks, making and supervising matches, organizing fundraising campaigns, recruiting volunteers, and conducting orientation and training sessions (Furano et al., 1993). This severely impacts the time that a case manager has available to supervise and support ongoing youth mentoring relationships. Furthermore, funding agencies often use the number of new matches generated as opposed to the longevity of matches as the measure of a program s success (Rhodes, 2002). Therefore, agencies tend to focus their limited resources into creating matches rather than sustaining matches already made (Rhodes, 2002). Due to the multiple demands placed on case managers, agencies need to develop strategies or procedures that help identify relationships at risk for premature termination or other problems. A primary task for case managers is to help adult volunteers gauge their feelings and expectations about the relationship as it progresses (Rhodes, 2002). This is important because research indicates that volunteers need to develop new motives once their initial ones are satisfied because their service no longer fulfills their original desires (Stukas & Tanti, 2005; Synder, et al., 2000). For example, a young adult decides to mentor because that individual wants to gain experience working with youth (career function), but once that volunteer accumulates some experience, the volunteer s commitment to mentor is fueled by a desire to improve or increase social connections or relationships (social function). Conversely, a volunteer s obsolete motives could be reignited by engaging in a different task (Stukas & Tanti, 2005). In the previous example, the mentor s initial career motives could be reignited by being offered the opportunity to mentor youth in a school or other specialized setting. The research described above provides a strong rationale for organizations to identify a volunteer s

29 PREDICTING INTENSITY OF CONTACT 21 expectations, and then to routinely assess if a volunteer s experiences are fulfilling these expectations, as they evolve. One challenge for assessing a volunteer s motives is social desirability, or the natural tendency to endorse attitudes and behaviors that portray one s self in a favorable light (Serow, 1991). Although mentors have been shown to experience a variety of emotions during their volunteer work, they may minimize emotions such as frustration, anger, and regret because of the perceived and often explicit expectation that they should be primarily concerned with helping the youth. Agencies need to provide volunteers with a way to share their feelings, needs, and motives that may be cognitively dissonant for them and that will minimize or reduce the fear of criticism or judgment. An agency s screening processes and procedures can also create obstacles for effectively recruiting and retaining volunteers. For many volunteers, the intake process is long, arduous, and often personally invasive (Furano, Roaf, Styles, & Branch, 1993, p. 10). The primary purpose for screening potential adult volunteers is to protect children from perpetrators and abuse. Case managers at Big Brothers and Big Sisters programs ask adult volunteers questions about their experiences with children, a description of their own family and upbringing, a description of lifestyle habits and values, and responses to hypothetical situations (Wilson & Beville, 2003). While these questions may help case managers in assessing and detecting potential adult perpetrators, the research suggests that the intensely personal nature of the interview can deter some adults from becoming volunteers. Fortunately, Clary et al. s (1998) Volunteer Functions Inventory (VFI) provides a tool that agencies can use to obtain information about the motives of volunteers in a less intrusive manner. Incorporating a paper and pencil tool into the interview process decreases the pressure

30 PREDICTING INTENSITY OF CONTACT 22 of a face-to-face interview and helps the volunteer provide more genuine and authentic information about their reasons for mentoring. The VFI assesses the personal and social motives, needs, goals, and functions an individual hopes to fulfill by volunteering. Clary et al. (1998) reported the results of three studies that demonstrated the VFI can be used to understand many concepts of volunteerism, especially recruitment and retention. First, the authors showed that motives, as measured by the VFI, predict the persuasiveness of different appeals to volunteers. Advertisements were more appealing when they more closely matched the prospective volunteer s motives. As such, the VFI is a functionally oriented measure that can help agencies accurately identify the motives of volunteers (Clary et al., 1998). Second, Clary and colleagues showed that satisfaction with volunteer activities depends on the match between an individual s motivational goals (as measured by the VFI) and the ability of the specific volunteer activities to fulfill those goals. In the third study, Clary and his colleagues showed that these satisfaction ratings were, in turn, associated with commitment to continue in the role of a volunteer. In conclusion, these results suggest that agencies could utilize functional theory to improve volunteer retention by developing activities that support evolving motives. Summary of the Research on Mentoring The critical role of enduring relationships in producing benefits for youth places a premium on retaining mentors. Unfortunately, almost half of newly formed mentoring relationships terminate within the first few months, which may actually be detrimental to the mentees (Rhodes, 2006; Rhodes & Grossman, 2002). Agencies, therefore, need to improve their ability to retain volunteers to decrease premature terminations and enable youth to experience the benefits provided by prolonged relationships.

31 PREDICTING INTENSITY OF CONTACT 23 Researchers have studied benefits experienced by adult mentors, investigated aspects of successful relationships, examined how to foster volunteer satisfaction, and identified variables that hinder the intensity of a volunteer s contact with a youth. The literature indicates that volunteers will persist as long as the rewards outweigh the costs. Other important findings reveal that the degree of perceived similarity between the individuals is a strong predictor of longevity (Madia & Lutz, 2004). However, these research findings have only had a moderate effect on a youth mentoring agency s ability to effectively recruit and retain mentors. Literature on youth mentoring indicates agencies can retain mentors more effectively by fostering mentor satisfaction, supporting development of relationships, and implementing practices that provide support, feedback, and recognition to mentors (Stukas & Tanti, 2005). Most importantly, because agencies have limited resources, research needs to provide efficient and cost-effective practices for targeting volunteers most likely to persist as mentors. The primary goal of this research is to help agencies improve recruitment and retention strategies by examining predictors of current volunteer engagement. Overview of the Study A group of new adult mentors at Big Brothers and Big Sisters program sites were surveyed using the Volunteer Functions Inventory (Clary et al., 1998) to examine variables that affect the intensity of a volunteer s contact in youth mentoring. This study attempted to expand on research conducted by Clary et al. (1998) and Madia and Lutz (2004), which revealed that individuals express stronger intentions to persist as volunteers if they perceive that their service fulfills initial expectations. However, these studies measured contact as a dichotomous outcome, asking participants if they intended to continue to volunteer. The results do not allow for any

32 PREDICTING INTENSITY OF CONTACT 24 conclusions to be made about the effect of a discrepancy between expectations and actual experience on the intensity of an adult volunteer s contact or engagement with the youth. Therefore, this study measured intensity of contact as a continuous variable by gathering information on hours of face-to-face contact between adult volunteer and youth during the month prior to completing the survey. Additionally, volunteers were surveyed about their expectations and experiences as a mentor, and asked to report on how often their case manager contacts them to provide support or supervision. In this study, adult volunteers are referred to as Bigs and the youths to whom they are matched are called Littles. Research Questions and Hypotheses The primary hypothesis of this study is that the degree of a perceived discrepancy between initial expectations and actual experience will affect the intensity of contact between a Big and Little. Intensity of contact is measured by the Big s report of face-to-face contact with their Little within the course of a month. Specifically, it is believed that a large degree of discrepancy between initial expectations and actual experience will decrease the intensity of contact with the youth mentee. Conversely, a low degree of discrepancy between initial expectations and actual experience should increase intensity of contact. A secondary objective of this study was to examine how frequency of match support provided by BBBS case managers moderates the relationship between an adult volunteer s expectation-experience discrepancy and intensity of contact. It is believed that high frequencies of match support should help minimize the effect of high discrepancies and facilitate an adult volunteer s engagement with the youth mentee. The last objective was to assess the relationship between volunteer contact and specific motivational factors, as measured by Clary et al. s (1998) Volunteer Functions Inventory. Prior research has revealed that different motives may predict success in different types of mentoring

33 PREDICTING INTENSITY OF CONTACT 25 activities (Stukas & Tanti, 2005, p. 241). In conclusion, the results from this study could be used by Big Brothers and Big Sisters to modify recruitment strategies to target persons who are likely to persist as volunteers, and to help case managers retain volunteers that might want to drop out by modifying that person s expectations so they experience higher levels of satisfaction. Such targeted recruitment and support could increase persistence and satisfaction, while decreasing the occurrence of premature terminations. Method Participants Participants in this study were adult volunteers from Big Brothers and Big Sisters community based programs that had been matched with a youth for 3-12 months. This group of mentors was selected because research indicated that mentoring relationships are at risk for premature termination, especially after the six-month mark (Rhodes & Grossman, 2002). Big Brothers and Big Sisters of America reported six-month retention rates of 84% as of December 2011 and 12-month retention rates of 64% as of November 2011 (D. Beturne, personal communication, April 4, 2012) Also, collecting data from volunteers up to 12 months postmatch provides a range of data on the mentor s impression of the work and sufficient variability to detect a potential effect of length of service on the relationship between an expectationexperience discrepancy and intensity of contact. The survey was not distributed to mentors matched for more than 12 months because it was assumed that their recall of initial expectations would be less vivid, and prolonged service would be associated with minimal discrepancies between expectations and experience.

34 PREDICTING INTENSITY OF CONTACT 26 Procedure Volunteers were recruited via an invitation (see Appendix B) distributed to eligible participants by each BBBS agency. This invitation included an informed consent document (see Appendix C) providing a link to a web-based version of the VFI. The survey was distributed to BBBS agencies throughout the United States. No demographic information was collected, but BBBS volunteers vary along multiple dimensions and characteristics such as age, sex, ethnicity, and socioeconomic status. Completed surveys were collected from the beginning of April 2013 until the end of August To maximize response rates, participants who completed the VFI and returned it within two weeks of the invitation were invited to enter into a drawing for a chance to win one of four $25 VISA gift cards. Please see Appendix D for the invitation to enter the prize drawing. Recruitment A total of eight BBBS agencies participated in this study. The Hampden County Office of Big Brothers and Big Sisters, located in Springfield, MA, distributed the survey to 30 eligible volunteers. Two other agencies, BBBS of Greater Flint (MI), and BBBS of Mississippi Valley (Davenport, IA) responded to a request posted on a listserv, distributing the survey invitation to 85 and 121 eligible volunteers, respectively. Subsequently, five additional BBBS agencies were sent direct inquiries seeking their assistance in surveying volunteers: BBBS of Southern Maine (Portland) sent out the survey to 29 volunteers; BBBS of Windham County (VT) distributed the survey to 100 volunteers; BBBS of Rockland County (NY) sent the survey out to 56 volunteers; BBBS of Central Indiana (Indianapolis, IN) sent out the survey to 264 eligible volunteers; and BBBS of Greater Seacoast in NH invited volunteers to participate through their

35 PREDICTING INTENSITY OF CONTACT 27 newsletter. At the time BBBS of Greater Seacoast, NH distributed the survey, they reported problems with their server and they were unable to provide a number of eligible volunteers. Based on information provided by seven out of the eight BBBS agencies (excluding BBBS of Greater Seacoast, NH), the survey was distributed to a total of 685 volunteers. The average number of participants surveyed by these seven agencies was 97 and this was used as an estimate of how many adult volunteers received the survey from BBBS of Greater Seacoast. Overall the survey was distributed to approximately 782 adult volunteers, from which 113 volunteers responded, yielding a response rate of 14%. Of the 113 survey respondents, 55 entered the prize drawing. The four winners were selected using a random number generator program and notified via . Measure An online survey was created using The Volunteer Functions Inventory developed by Clary et al. (1998) and administered to participants to assess discrepancies between their initial motives for volunteering and their actual experience. Subscale scores were calculated based on the scoring sheet provided by Clary, Stukas, and Synder (personal communication, September, ). Research on the VFI has demonstrated that each of the VFI s six subscales distinctively capture different motivations and functions for volunteerism (Clary et al., 1998). Clary et al. administered the VFI to volunteers from five different organizations and demonstrated adequate internal consistency of the VFI with Cronbach alpha scores ranging from.80 to.89 for each of the six subscales. Clary and his colleagues also conducted a factor analysis to assess the instrument s discriminant validity. Results revealed six components with eigenvalues greater than 1.0, which demonstrates that the six VFI scales Clary and his colleagues created using functional theory are distinct constructs. Last, Clary et al. administered the VFI

36 PREDICTING INTENSITY OF CONTACT 28 and the 17 functional benefit questions to 61 older volunteers at a community hospital. These 17 questions are designed to assess a volunteer s perception of how well her or his service has satisfied his or her motives. Clary et al. (1998) assessed the internal validity of these questions by computing Cronbach s alpha for each subscale and reported scores ranging from.75 to.89. In sum, the VFI has good psychometric properties for assessing volunteer motives across a range of populations. The VFI contains two sections. Section One of the VFI contains 30 items that assesses a volunteer s motivational preferences, while Section Two contains 17 functional benefit questions. The VFI offers separate scores that correspond to six motivating functions: values, understanding, social, career, protective, and enhancement. The definition for each function is as follows: (a) Value function- volunteerism offers an individual the opportunity to express values associated with altruism and other humanistic beliefs; (b) Understanding function- volunteering allows individuals the opportunity to be exposed to different experiences and share knowledge, skills, and abilities that they might not otherwise use; (c) Social function- volunteer work offers the opportunity to spend time with friends or other important people or engage in an activity that other people value; (d) Career function- volunteerism offers a volunteer the opportunity to learn, practice, and even maintain skills needed to perform a particular job; (e) Protective function- volunteer work helps people to escape negative feelings or even help one deal with one s own personal problems; and (f) Enhancement function- people engage in volunteer work because it helps enhance mood, affect, self-esteem, and promotes personal growth. (Clary et al., 1998, pp )

37 PREDICTING INTENSITY OF CONTACT 29 Modifications were made to the VFI for the purposes of this research. This study only utilized Section One of the VFI because the difference in the number of questions between Section One and Section Two would make it difficult to calculate a valid discrepancy score. In addition, it was decided that having participants respond to similar questions when rating how well their experience fulfilled initial expectations would improve the validity and reliability of responses. Instructions for Form A were modified to request that respondents think back to when they first started volunteering and indicate the degree to which items reflect an initial expectation they had about how they would benefit from being a mentor. Instructions for Form B were changed to ask respondents to indicate the extent to which their experiences as a mentor matched up with their initial expectations. In addition, wording on many items of Form B were modified to ask participants to estimate how well their experience fulfilled expectations. Respondents were instructed to rate the items on both forms using a 5-point Likert scale with all points labeled. The rating scales were shortened from seven points to five and all points were labeled to improve the quality of data provided by respondents. Research indicates that the validity and reliability of data tends to be strongest when rating scales contain between five and seven points that are all verbally labeled (Krosnick & Fabrigar, 1997). Description and Measurement of Variables Calculation of expectations-experience discrepancy. Respondents completed two versions of Section One of the VFI, as described above. Discrepancy scores were generated using a similar approach described by Madia and Lutz (2004); that is, by computing a difference score for each subscale from Forms A and B, such that positive scores indicate that the volunteer s experience did not fulfill initial expectations, and negative scores indicate that the volunteer s experience provided more benefits than initially expected.

38 PREDICTING INTENSITY OF CONTACT 30 Intensity of contact. This variable was measured on a continuous scale, by asking respondents to report the number of hours of face-to-face contact with their Little during the month prior to when they completed the survey. Frequency of match support. This variable is defined by the amount of contact between a BBBS case manager and adult volunteer. Respondents were asked to answer a multiple choice question to indicate how often they had contact with a BBBS case manager. The three response choices were: (a) every two weeks, (b) once a month, and (c) every three months. Please refer to Table 1 for a description of the study s variables, hypotheses, and analyses.

39 PREDICTING INTENSITY OF CONTACT 31 Table 1 Description of Variables, Hypotheses, and Analyses Variable Hypothesis Analysis Predictor Variable Criterion Variable Discrepancy score Hours of faceto-face contact in past month Lower discrepancy scores predict a higher intensity of contact Regression analysis Frequency of match support & Discrepancy score Hours of faceto-face contact in past month More frequent match support will weaken the relationship between discrepancy score and intensity of contact. Moderation analysis Initial Expectation VFI subscale scores (Form A) Hours of faceto-face contact in past month Volunteers who receive certain benefits will have a higher intensity of contact Correlation Results Data Preparation Prior to analyzing the data, the entire data set was inspected to identify any outliers and ensure participants responded to all items. First, seven cases were filtered out that were missing values on the primary criterion variable, hours of face-to-face contact. This reduced the total number of cases used in the analyses to 106. Next, two cases were identified as potential outliers with these two respondents reporting having spent 40 and 50 hours of face-to-face contact with

40 PREDICTING INTENSITY OF CONTACT 32 their Little. While these scores are very high, they are not unrealistic when considering data was gathered during the summer months when school vacation occurs and it is not unusual for a Big and Little matched for close to a year to engage in overnight activities such as going on a camping trip or visiting other places of mutual interest; thus, these scores were retained in the dataset. Then, the predictor variable frequency of match support was collapsed from three to two levels, because only five participants reported receiving match support every two weeks. Those five were combined with the 67 participants who reported receiving monthly supervision, resulting in two supervision levels: at least once per month (n=72), and once every three months (n=34). The next step that was taken in preparing the data for analysis was computing VFI subscale scores for initial expectations (Form A) and actual experience (Form B). Then, discrepancy scores were calculated for each subscale by subtracting Form B scores from Form A scores. Last, an overall discrepancy score for the VFI Total Score was calculated using the same method described above. Distributions of all variables were examined, and are presented in Appendices E (hours of face-to-face contact, frequency of match support, and total discrepancy score), F (VFI subscale initial expectation scores from Form A), G (VFI subscale experience scores from Form B), and H (discrepancy scores for each VFI subscale). Descriptive statistics for the focal variables in this study are presented in Table 2.

41 PREDICTING INTENSITY OF CONTACT 33 Table 2 Descriptive Statistics for Hours of Face-to-Face Contact and Discrepancy Score Variable M SD 95% Cl Hours of Face to Face Contact Overall Discrepancy Score n= [7.20, 9.80] [-0.17, -0.01] Predicting Intensity of Contact Using a discrepancy between initial expectations and experience to predict intensity of contact. First, the data was visually inspected using a scatter plot (see Figure 1). BBBS volunteers in this sample reported low levels of discrepancy between experience and expectations, with 86% of scores falling within -0.5 and 0.5, on a scale of In addition, more than half of the respondents reported negative discrepancy scores, indicating that their actual experience exceeded initial expectations. Of note, Figure 1 shows that in some instances participants who reported positive discrepancy scores reported a higher number of hours of face-to-face contact compared to participants who reported negative discrepancy scores.

42 PREDICTING INTENSITY OF CONTACT 34 Figure 1 Scatter plot of Discrepancy Score and Hours of Face-to-Face Contact The graphic in Figure 1 also shows hours of face-to-face contact spanned from as little as 2 hours to as many as 50 hours. The variability in hours of face-to-face contact is surprising because the low end of the observed range is well below the minimum expectation set by BBBS of eight hours per week, while the two outliers at the high end are quite extreme. An inspection of the scatter plot indicates a flat profile and it seems fairly clear that the two outliers are not skewing the relationship between the discrepancy score and intensity of contact. Next, distributions of the variables to be entered into the regression model were examined for approximation to a normal distribution and were considered compliant with the statistical assumptions. Although raw VFI subscale scores for Career, Social, and Protective subscales (Appendices G and H) deviated from a normal distribution, the associated discrepancy scores did

43 PREDICTING INTENSITY OF CONTACT 35 not. All subscale outcome scores were given equal weight in the equation because there is no evidence suggesting one variable is more influential than another on intensity of contact. The regression analysis confirmed that the degree of discrepancy between initial expectations and actual experience did not reliably predict intensity of contact (B = -.39; p =.84; see Table 3).

44 PREDICTING INTENSITY OF CONTACT 36 Table 3 Using a Discrepancy Score to Predict Intensity of Contact Intensity of Contact Variable B α 95% CI Discrepancy Score [-3.64, 2.86] Examining the relationship between frequency of match support, discrepancy score, and intensity of contact. Initially, a moderation analysis was planned to test the effect of frequency of match support on the relationship between discrepancy score and intensity of contact. However, given that no visible or statistically significant relationship between discrepancy score and intensity of contact emerged, a more exploratory examination of associations between frequency of match support, intensity of contact, and discrepancy score was undertaken. Mean scores for intensity of contact and discrepancy, by levels of match support, are presented in Table 4.

45 PREDICTING INTENSITY OF CONTACT 37 Table 4 Two Level Match Support and Associated Discrepancy Scores and Hours of Face-to-Face Contact Discrepancy Score Intensity of Contact M SD 95% CI M SD 95% CI Frequency of Match Support At least monthly [-0.23, -0.03] [6.93, 10.07] Every 3 months [-0.13, 0.12] [6.01, 11.05] A breakdown of responses shows that 68% of participants reported receiving match support at least monthly, while 32% reported receiving match support once every three months. It was believed that frequent match support in this sample would be associated with more intense contact by resolving expectation-experience discrepancies and increasing accountability in meeting expectations for contact; however, the correlation matrix among these variables did not reveal any significant associations (see Table 5).

46 PREDICTING INTENSITY OF CONTACT 38 Table 5 Correlation Matrix for Frequency of Match Support, Discrepancy Score, and Intensity of Contact Variable Frequency of match support 2. Discrepancy total 3. # of hours of face-to-face contact Analyzing the effect of initial expectations on intensity of contact. The last objective of this research was to identify relationships between specific motivational factors, as measured by the VFI subscales, and volunteer engagement. The six motivational subscale scores from Form A represent the predictor variables and intensity of contact is the criterion variable. The correlations were calculated for each the six VFI subscales, the total mean score for Form A, and intensity of contact. The results revealed that the six VFI subscales and the total score for Form A were not significantly correlated with intensity of contact (see Table 6).

47 PREDICTING INTENSITY OF CONTACT 39 Table 6 Correlations between Initial Expectations and Intensity of Contact Variable Value 2. Understand.44** 3. Protective 4. Social 5. Enhancement 6. Career.20*.60**.11.32**.27**.22*.61**.76**.29**.06.48**.39**.36**.52** 7. Total Form A ** -.31** * Intensity of Contact * Correlations were significant at p <.05 ** Correlations were significant at p <.01 Discussion Summary of Findings The results of this research did not reveal any reliable predictors for intensity of contact for adult volunteers in Big Brothers and Big Sisters community-based programs. Survey data was collected from 113 volunteer BBBS mentors in eight states. The primary hypothesis in this study was that the volunteer s experienced degree of discrepancy between initial expectations and actual experience would affect intensity of contact with the youth. However, the results suggest that the effects of a discrepancy between a volunteer s expectations and experience on intensity of contact are minimal. Secondary hypotheses involved examining how frequency of match support affects discrepancy score and intensity of contact. It was posited that more

48 PREDICTING INTENSITY OF CONTACT 40 frequent match support would help lessen the discrepancy between initial expectations and actual experience and help to facilitate volunteer engagement. Results of the analysis revealed that frequency of match support showed no association with intensity of contact or the discrepancy score. The last objective of this study was to identify relationships between specific motivational factors as measured by the VFI subscales and intensity of contact. This analysis indicated none of the six VFI subscales were related to intensity of contact. Even though none of the analyses provided statistically significant results, there are some notable findings and they will be discussed in the next section. Explanation of Findings Using a discrepancy between initial expectations and experience to predict intensity of contact. The vast majority of respondents experienced a minimal discrepancy between initial expectations and experience. This finding could be attributed to procedures BBBS agencies already use, such as the lengthy and thorough volunteer screening process, matching a Big and Little based on compatibility of interests and goals, and ongoing match support that helps to resolve or minimize any expectation-experience discrepancies. Therefore, it s possible that the procedures used by BBBS agencies are effective in minimizing discrepancies between initial expectations and actual experience, eliminating its effect on intensity of contact. Another possible explanation could be that BBBS mentors who experienced high levels of discrepancy might have dropped out before even reaching the three month threshold for participation in this study. At the same time, the number of hours of face-to-face contact showed very substantial variability, ranging from 2 hours to 50 hours, which must be attributed to factors not explored in this study, such as relocation of either Big or Little, changes in a Big or Little s availability, or lack of a supportive parent.

49 PREDICTING INTENSITY OF CONTACT 41 Cognitive dissonance theory could also explain the limited range in discrepancy scores among these active volunteers. Cognitive dissonance theory proposes that when an individual experiences conflicting attitudes, beliefs, or behaviors it produces a feeling of discomfort that causes the individual to change or alter one of the attitudes, beliefs, or behaviors to reduce the discomfort (Myers, 1999). Dissonance theory would predict that volunteers, who had made the decision to remain active, would tend to minimize their awareness or reporting of unmet expectations. BBBS agencies could apply the theory to increase volunteer retention and persistence through a process of effort justification (McLeod, 2008). For example, case managers can emphasize that the volunteer helped to facilitate even the smallest changes in the youth, and the volunteer can use that information to justify that their efforts have been worthwhile, which will reduce dissonance caused by feeling ineffective or unhelpful so the volunteer persists in their role. This could help to minimize any discrepancies between the volunteer s expectations and experience. The lack of any relationship between a volunteer s discrepancy score and intensity of contact also raises some questions about factors that influence the volunteer s commitment to mentoring. Self-justification theory offers an explanation of how people resolve cognitive dissonance and its effects on commitment to an activity. Self-justification theory proposes that, when a person encounters cognitive dissonance, that person tends to justify the behavior and deny any negative feedback associated with the behavior (Staw, 1976). For example, an active and satisfied volunteer who fails to meet the minimum expectation for contact with their mentee might rationalize the behavior by reminding themselves that their mentee is doing well or denying any negative effects on the youth. These justifications help the volunteer minimize any dissonance and continue their commitment to their role.

50 PREDICTING INTENSITY OF CONTACT 42 Examining the relationship between frequency of match support, discrepancy score, and intensity of contact. One possible reason that frequency of match support and discrepancy scores was not related in this study could be due to a relatively restricted range in discrepancy scores among a sample of active volunteers. It s possible that the effect of frequency of match support would be more noticeable if discrepancy scores and intensity of contact from active volunteers were compared to data from dropouts. Analyzing the effect of initial expectations on intensity of contact. In this sample, none of the six VFI subscales scores from Form A were related to intensity of contact. The lack of significant correlations between any of the VFI subscales and intensity of contact is surprising because research has shown that people engage in volunteer work to fulfill specific motives. Overall, these results suggest that understanding an individual s expectations about the volunteer role may be more useful when recruiting volunteers, but minimally helpful in sustaining involvement. Limitations of the Study The results do raise questions about a few aspects of the study s design. First, as mentioned above, including data from dropouts might have enhanced the exploration of the role of experience-expectation discrepancies and match support in understanding mentor persistence. The challenges of collecting data from dropouts were daunting, but may have hampered the study s results. In addition, a cross-sectional design could have contributed to a lack of variability in discrepancy scores and weakened the associations between variables in a few ways. First, a volunteer s recall of initial expectations becomes less accurate the longer they have been involved. Second, it s likely that surveying only active volunteers introduced a self selection

51 PREDICTING INTENSITY OF CONTACT 43 bias (Olsen, 2008), reducing variability in discrepancy scores. Self selection bias would be present if active, satisfied volunteers were more likely to complete the survey than less satisfied volunteers. This problem occurs in most survey samples because respondents have the option to decide whether or not they want to complete the survey (Olsen, 2008). Lastly, it s possible that the method used to estimate intensity of contact resulted in measurement error that attenuated the potential for prediction. Ongoing monitoring of contact logs and computation of an average, rather than a single snapshot measurement of intensity of contact, might enhance the sensitivity of these analyses. Therefore, using a longitudinal design might enhance the power of this line of inquiry. Comparison of Results with Past Research Previous studies by Clary et al. (1998) and Madia and Lutz (2004) reported that factors such as volunteer satisfaction and expectation-experience discrepancies were related to an expressed intention to continue to volunteer. One difference between this study and the two described previously is that neither of the latter explored the relationship of these variables with current involvement, nor was any follow up conducted to see if participants followed through on those intentions. This study attempted to take that research one step further and identify variables related to current involvement. In this study, variables shown by Clary et al. and Madia and Lutz to be predictive of intentions to engage in a volunteer activity did not appear to be reliable predictors of current volunteer engagement. Areas for Future Research The findings from this study point to areas that warrant further exploration in future research. In light of the limited range of discrepancy between expectations and experience of mentoring in the sample recruited for this study, one obvious question is whether BBBS mentors

52 PREDICTING INTENSITY OF CONTACT 44 who experienced high levels of discrepancy dropped out before even reaching the 3-month threshold for participation in this study. This question would best be answered by prospectively following a group of BBBS volunteers through their first 3-6 months of engagement. It must be noted, though, that BBBS research cited earlier in this paper indicates that dropout prior to 6 months accounts for only 16% of volunteer attrition (D. Beturne, personal communication, April 4, 2012), which would seem to suggest that, if the expectation-experience discrepancy is only influential within the first 3-6 months of BBBS service, it is unlikely to exert high leverage on overall volunteer retention for BBBS. If the expectations-experience discrepancy is not a major determinant of intensity of contact between BBBS mentors and their mentees, what other factors might be driving the variation in that intensity observed in this study? Possibilities include changes of residence or other life circumstances that interfere with BBBS participation for Littles who are drawn disproportionately from populations with relatively chaotic lifestyles, or lack of sufficient parental support for sustaining involvement. The current study focused on the motivation of the Big, and other possibilities residing on their side of the relationship remain worthy of further study, though perhaps using more exploratory, qualitative methods. One area within the potential control of BBBS agencies would be features of the match support provided to their mentors, though frequency of match support was not associated with intensity of contact between Bigs and Littles in the current study. Qualitative strategies could be used to explore how specific obstacles affect mentoring relationships (e.g., difficulties coordinating meetings due to competing demands outside the relationship or problems finding, planning, and engaging in activities). Such findings might illuminate potential refinements to match support provided by BBBS agencies. Given chronic resource challenges faced by BBBS

53 PREDICTING INTENSITY OF CONTACT 45 agencies, match support is at high risk of erosion, and it would be helpful to know as much as possible about critical ingredients for sustaining mentor commitment and engagement. Finally, perhaps the assumptions underlying BBBS s 8 hour per month minimum expectation for Big-Little contact bear further examination. In particular, is there some minimal frequency threshold for Littles to experience benefit, or is it possible to derive benefit from less frequent contact? If so, what circumstances are associated with successful but less frequent Big-Little relationships, and what ingredients are important to their success? Perhaps those Bigs reporting as few as 2 hours per month of contact in this study, don t represent failed mentoring relationships after all. There is precedent in the resilience literature for adult child relationships of limited scope or duration, exerting seemingly disproportionate leverage on the child s perception of his or her prospects in life (Scales, 2006). If intensity of contact between the adult mentor and mentee is not a primary determinant of the benefits a youth experiences from the relationship, the effort to assert minimum requirements for face-to-face contact may be misplaced, and BBBS agencies could shift their focus to exploring and understanding the intrapersonal determinants and psychological processes that influence a volunteer s decision to engage and commit to a particular activity. Conclusions In conclusion, mentoring is designed to have adults help youth learn and develop skills and competencies to be successful and promote self-esteem. Research has shown youth mentoring to be a very helpful intervention in improving the lives of young people. However, agencies have had trouble meeting the demand for youth mentoring services. This study highlights the difficulties agencies face in meeting the demands for this service because it provided no clear variables or factors that reliably predict or are related to current volunteer

54 PREDICTING INTENSITY OF CONTACT 46 engagement. It is vital that agencies continue to increase their knowledge of how to recruit and retain volunteers so more youth can experience the benefits of this intervention.

55 PREDICTING INTENSITY OF CONTACT 47 References Anderson, J.C., & Moore, L.F. (1978). The motivation to volunteer. Journal of Voluntary Action Research, 7, Allen, T. D. (2003). Mentoring others: A dispositional and motivational approach. Journal of Vocational Behavior, 62, Big Brothers Big Sisters. (2014a). 110 years of history. Retrieved from c.9iili3ngkhk6f/b /k.e56c/starting_something_since_1904.htm Big Brothers Big Sisters. (2014b). Whom we serve. Retrieved from c.9iili3ngkhk6f/b /k.7035/something_for_everyone.htm Brown-Wright, D., Dubick, R., & Newman, I. (1997). Graduate assistant expectation and faculty perception: Implications for mentoring and training. Journal of College Student Personnel, 38, Bogat, G. A., & Liang B. (2005). Gender in mentoring relationships. In D.L. DuBois & M. L. Karcher (Eds.), Handbook of youth mentoring (pp ). Thousand Oaks, CA: Sage Publications. Bussell, H., & Forbes, D. (2002). Understanding the volunteer market: The what, where, who, and why of volunteering. International Journal of Nonprofit and Voluntary Sector Marketing, 7, Clary, E. G., Snyder, M., & Ridge, R. D. (1992). Volunteer motivations: A functional strategy for the recruitment, placement, and retention of volunteers. Nonprofit Management & Leadership, 2,

56 PREDICTING INTENSITY OF CONTACT 48 Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas, A. A., Haugen, J., & Miene, P. (1998). Understanding and assessing the motivation of volunteers: A functional approach. Journal of Personality and Social Psychology, 74, DuBois, D. L. & Karcher, M. J. (2005). Youth mentoring: Theory, research, and practice. In D. L. DuBois & M. L. Karcher (Eds.), Handbook of youth mentoring (pp. 2 11). Thousand Oaks, CA: Sage Publications. Eagly, A. H., & Crowley, M. (1986). Gender and helping behavior: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100(3), Eisner, D., Grimm, R. T., Maynard, S., & Washburn, S. (2009). The new volunteer workforce. Stanford Social Innovation Review, Retrieved from /articles/entry/the_new_volunteer_workforce. Furano, K, Roaf, P. A., Styles, M. B., & Branch, A. Y. (1993). Big Brothers/ Big Sisters: A study of program practices (NCJ Number ). Retrieved from National Criminal Justice Reference Service website: /Abstract.aspx?id= Grube, J. A., & Piliavin, J. A. (2000). Role identity, organizational experiences, and volunteer persistence. Personality and Social Psychology Bulletin, 26, Hays, R.B., & Oxley, D. (1986). Social network development and functioning during a life transition. Journal of Personality and Social Psychology, 50(2), Klein, N. A., & Sondag, K. A. (1994). Understanding volunteer peer health educators motivations: Applying social learning theory. Journal of American College Health, 43, 5 14.

57 PREDICTING INTENSITY OF CONTACT 49 Krosnick, J.A., & Fabrigar, L.R. (1997). Designing rating scales for effective measurement in surveys. In L. Lyberg, P. Blemer, M. Collins, E., de Leeuw, C Dippo, N. Schwarz, & D. Trewin (Eds.), Survey measurement and process quality ( ). New York, NY: John Wiley. Larkin, E., Sadler, S. E., & Mahler, J. (2005). Benefits of volunteering for older adults mentoring at-risk youth. Journal of Gerontological Social Work, 44, Madia, B.P., & Lutz, C.J. (2004). Similarity, positive relationship characteristics, and mentors' expressed intention to remain in Big Brothers/Big Sisters Programs. Journal of Applied Social Psychology, 34, McDonald-Hart, D. (2002). The impact on mentors maintaining quality relationships with low-achieving elementary students/mentees grades one through six. (Doctoral dissertation) Retrieved from Proquest. (UMI No ). McLeod, S. A. (2008). Cognitive dissonance. Retrieved from /cognitive-dissonance.html Myers, D. G. (1999). Behavior and attitudes. In Social Psychology (6 th ed.) (pp ). New York: McGraw-Hill College. National Partnership for Mentoring. (2005). A snapshot of the current state of mentoring. Retrieved from Olsen, R. (2008). Self-selection bias. In P.J. Lavrakas (Ed.). Encyclopedia of Survey Research Methods. (pp ). Available from /ehost/detail?sid=353ea97e-85a8-4ee5-956d-659b9e50d4d0@sessionmgr198&vid=1# db=e000xna&an= Omoto, A. M., & Snyder, M. (1995). Sustained helping without obligation: Motivation,

58 PREDICTING INTENSITY OF CONTACT 50 longevity of service, and perceived attitude change among AIDS volunteers. Journal of Personality and Social Psychology, 68, Penner, L. A. (2002). Dispositional and organizational influences on sustained volunteerism: An interactionist perspective. Journal of Social Issues, 58, Penner, L. A., & Finklelstein, M. A. (1998). Dispositional and structural determinants of volunteerism. Journal of Personality and Social Psychology, 74, Penner, L. A., Fritzsche, B. A., Craiger, J. P., & Freifield, T. R. (1995). Measuring the prosocial personality. In J. Butcher & C. D. Speilberger (Eds.), Advances in personality assessment. (pp ). Hillsdale, NJ: Erlbaum. Philip, K., & Hendry, L. B. (2000). Making sense of mentoring or mentoring making sense? Reflections on the mentoring process by adult mentors with young people. Journal of Community & Applied Social Psychology, 10, Puzzanchera, C., & Adams, B. (2011). Juvenile Arrests 2009 (Research Report No. NCJ ). Retrieved from the Office of Juvenile Justice and Delinquency Prevention Rhodes, J. E. (2002). Stand by Me: The Risks and Rewards of Mentoring Today s Youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. L. Karcher (Eds.), Handbook of youth mentoring (pp ). Thousand Oaks, CA: Sage Publications. Rhodes, J. E. (2006). Strategies for recruiting and retaining mentors. Retrieved from Rhodes, J. E., Bogat, G. A., Roffman, J., Edelman, P., & Galasso, L. (2002). Youth mentoring

59 PREDICTING INTENSITY OF CONTACT 51 in perspective: Introduction to the special issue. American Journal of Community Psychology, 30, Rhodes, J. E., & Grossman, J. B. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, Scales, P.C. (2006). The world of adults today: Implications for positive youth development. In E. G. Clary & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors (pp ). New York, NY: Springer. Serow, R. C. (1991). Students and voluntarism: Looking into the motives of community service participants. American Educational Research Journal, 28, Sipe, C. L. (2002). Mentoring programs for adolescents: A research summary. Journal of Adolescent Health, 31, Staw, B.M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance 16(27), Stukas, A. A., Daly, M., & Clary, E. G. (2006). Lessons from research on volunteering for mobilizing adults to volunteer for positive youth development. In E. G. Clary & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors (pp ). New York, NY: Springer. Stukas, A. A., & Tanti, C. (2005). Recruiting and sustaining volunteer mentors. In D. L. DuBois & M. L. Karcher (Eds.), Handbook of youth mentoring (pp ). Thousand Oaks, CA: Sage Publications.

60 PREDICTING INTENSITY OF CONTACT 52 Synder, M., Clary, E. G., & Stukas, A. A. (2000). The functional approach to volunteerism. In G. R. Maio & J. M. Olson (Eds.), Why we evaluate functions of attitudes (pp ). Mahwah, NJ: Lawrence Erlbaum Associates. Tierney, J. P., Grossman, J. B., & Resch, N. (1995). Making a difference: An impact study of Big Brothers/Big Sisters. Public/Private Ventures, Philadelphia, PA. U.S. Census Bureau. (2012). Statistical Abstract of the United States. Retrieved from Westhues, A., Clarke, L., Watton, J., & St. Claire-Smith, S. (2001). Building positive relationships: An evaluation of process and outcomes in a big sister program. The Journal of Primary Prevention, 21, Wilson, J. (2000). Volunteering. Annual Review of Sociology, 26, Wilson, M.K., & Beville, B.A. (2003). Preemployment and volunteer screening: Reducing the risk of child sexual abuse for social service agencies. Families in Society, 84(2), Wilson, P. J., & Johnson, B. W. (2001). Core virtues for the practice mentoring. Journal of Psychology and Theology, 29,

61 PREDICTING PERSISTENCE OF MENTORS 53 Appendix A Permissions INVOICE NO. N/A Federal Tax I.D Date: September 30, 2014 IN MAKING PAYMENT REFER TO THE ABOVE INVOICE NUMBER Matthew Stevens, Department of Clinical Psychology Antioch University-New England 40 Avon Street Keene, NH APA Permissions Office 750 First Street, NE Washington, DC permissions@apa.org Fax: IF THE TERMS STATED BELOW ARE ACCEPTABLE, PLEASE SIGN AND RETURN ONE COPY TO APA. RETAIN ONE COPY FOR YOUR RECORDS. PLEASE NOTE THAT PERMISSION IS NOT OFFICIAL UNTIL APA RECEIVES THE COUNTERSIGNED FORM AND ANY APPLICABLE FEES. Request is for the following APA-copyrighted material: Scale content Table 1 (adapted), p. 1520, from Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas, A. A., Haugen, J., & Miene, P. (1998). Understanding and assessing the motivations of volunteers: A functional approach. Journal of Personality and Social Psychology, 74(6), doi: / For the following use: Non-Commercial Research or Educational Use in: a) thesis or dissertation research (such as data collection or surveys) via an online password-protected web site and/or in hardcopy format; and b) print and/or digital versions of the final thesis or dissertation document provided that digital distribution is limited to non-commercial, secure and restricted web site(s). File: Stevens, Matthew (author) Permission is granted for the nonexclusive use of APA-copyrighted material specified on the attached request contingent upon fulfillment of the conditions indicated below:

62 PREDICTING PERSISTENCE OF MENTORS The fee is waived. 2. The reproduced material must include the following credit line: Copyright 1998 by the American Psychological Association. Reproduced [or Adapted] with permission. The official citation that should be used in referencing this material is [list the original APA bibliographic citation]. 3. For all online use: (a) The following notice must be added to the credit line: No further reproduction or distribution is permitted without written permission from the American Psychological Association; (b) the credit line must appear on the first screen on which the APA content appears; and (c) the APA content must be posted on a secure and restricted web site. This agreement constitutes permission to reproduce only for the purposes specified on the attached request and does not extend to future editions or revisions, derivative works, translations, adaptations, promotional material, or any other formats or media. Permission applies solely to publication and distribution in the English language throughout the world, unless otherwise stated. No changes, additions, or deletions to the material other than any authorized in this correspondence shall be made without prior written consent by APA. This permission does not include permission to use any copyrighted matter obtained by APA or the author(s) from other sources that may be incorporated in the material. It is the responsibility of the applicant to obtain permission from such other sources. ACCEPTED AND AGREED TO BY: PERMISSION GRANTED ON ABOVE TERMS: Electronic signature: Karen A. Thomas Applicant for the American Psychological Association Date Date September 30, 2014 I wish to cancel my request for permission at this time.

63 PREDICTING PERSISTENCE OF MENTORS 55 Appendix B Recruitment Hello, my name is Matthew Stevens. I ve been a volunteer with Big Brothers/Big Sisters in Springfield, MA, for 9 years. I m also a clinical psychology doctoral student, doing research to examine factors that predict intensity of contact in adult volunteers. I am reaching out to you to see if you would be willing to complete an online survey that should take about 10 minutes of your time. Your answers will remain anonymous. The goal of this research is to help mentoring programs improve recruitment and retention of volunteers. Mentoring is a valuable service for the children and adolescents with whom we work. Unfortunately, mentoring agencies struggle to effectively recruit and retain volunteers, limiting their ability to provide this service to the many children in need. I hope to learn how to help Big Brothers/Big Sisters support and retain more mentors like yourself. If you decide to participate in this research, you will be offered the chance to win one of four $25 VISA gift cards. Your odds of winning will be no less than 4 in 25. For more details about the survey, and to have an opportunity to complete it if you wish, please go to the next page. Thank you for considering helping me with this project.

64 PREDICTING PERSISTENCE OF MENTORS 56 Appendix C Participant Informed Consent Factors that Effect Contact between Adults and Youth in Mentoring Relationships Dear BBBS volunteer mentor, Purpose of Project I am asking for your help in learning about factors that affect contact between adults and children in mentoring relationships. My goal is to help agencies keep adults and youth connected for long periods of time. Mentoring relationships are most helpful when they last at least six months. If you choose to participate You will be asked to fill out 2 forms. The first form contains 30 items. This form asks you to rate how well each item represents an important expectation you had about how you would benefit by becoming a mentor. The second form contains 32 items. This form will ask you to indicate how well your experience has fulfilled your expectations. It should take about 10 minutes to complete both forms. Benefits of this project This research may help inform BBBS agencies about how to better support adult mentors. You may become more aware of the fit between your initial expectations of being a mentor, and your experience as a mentor. You will also have the opportunity to enter a drawing for one of four VISA gift cards. If you begin the survey within two weeks from (enter date of ), you can enter a drawing for one of four $25 gift cards. I am trying to recruit mentors to complete this survey. Your odds of winning will be no less than 4 in 100. Risks of this project We do not forsee any risks to you from participating in this project, unless thinking about your expectations of being a mentor leads you to feel dissatisfied with your experience. If you find this happening to you, we encourage you to speak with your case manager about it. We will protect your privacy This survey will not collect any information about who you are - there is no way for anyone, including me, to know who provided which answers. You will be directed to a different survey if you decide to enter the prize drawing, so that I can know and notify winners of the prizes. Your participation is completely voluntary No one at your BBBS agency will know whether you participated. You can stop responding to the survey questions at any time. You do not need to finish the surveys to be eligible for the prize drawing you just need to skip to the last page and click on the link to the drawing. If you wish to complete the survey please link on paste the link below into your browser.

65 PREDICTING PERSISTENCE OF MENTORS 57 Please feel free to contact me with any questions using the information below. Any questions about your rights as a research participant can directed to 1 of these 2 individuals: Dr. Catherine Clarke, Ph.D. Chair of the Antioch University New England IRB Stephen P. Neun Vice President for Academic Affairs Thank you, Matthew Stevens

66 PREDICTING PERSISTENCE OF MENTORS 58 Appendix D Sweepstakes Drawing Please complete the question below to be entered into the drawing to win 1 of 4 $25 VISA gift cards. Your odds of winning will be no less than 4 in 100. Please provide me with your address so I can contact you if you win a VISA gift card.

67 PREDICTING PERSISTENCE OF MENTORS 59 Appendix E Distributions of Primary Research Variables Figure F1: Frequency of hours of face-to-face contact Figure F2: Frequency of Match Support

68 PREDICTING PERSISTENCE OF MENTORS 60 Figure F3: Total Discrepancy Score

69 PREDICTING PERSISTENCE OF MENTORS 61 Appendix F Graphs of Initial Expectations for the Six VFI Subscale Scores (Form A) Figure G1: Career Function Figure G2: Social Function

70 PREDICTING PERSISTENCE OF MENTORS 62 Figure G3: Values Function Figure G4 : Understanding Function

71 PREDICTING PERSISTENCE OF MENTORS 63 Figure G5: Enhancement Function Figure G6: Protective Function

72 PREDICTING PERSISTENCE OF MENTORS 64 Appendix G Graphs of Experience Rating Scores for the Six VFI Subscales (Form B) Figure H1: Career Function Figure H2: Social Function

73 PREDICTING PERSISTENCE OF MENTORS 65 Figure H3: Values Function Figure H4: Understanding Function

74 PREDICTING PERSISTENCE OF MENTORS 66 Figure H5: Enhancement Function Figure H6: Protective Function

75 PREDICTING PERSISTENCE OF MENTORS 67 Figure I1: Career Function Appendix I Graphs of Discrepancy Rating Scores for the Six VFI Subscales Figure I2: Discrepancy Function